PROJECT SUMMARY Up to 40% of children with epilepsy do not respond to available antiepileptic drugs (AEDs), and identifying genes for outcome and AED response will provide critical insight into underlying pathways. Genotyping can readily be performed on tens of thousands of patients, but phenotyping remains a largely manual task. Electronic medical records (EMR) have been implemented over the last two decades. This readily available data source has enabled large studies linking EMR and biorepositories to identify novel disease genes. However, EMR data have not been used in epilepsy genetic studies so far. The long-term goal is to better understand how genetic changes in childhood epilepsies predict specific phenotypes, medication responses, and outcomes. The overall objective of this study is to detect genetic risk factors by utilizing EMR data to identify new biological mechanisms. The central hypothesis is that while the complexity of the age-related clinical patterns of the childhood epilepsies creates a major obstacle in generating universally applicable phenotyping algorithms, alternative methods leveraging the similarity of the clinical disease course and medication trajectory can be used to identify causative genetic variants associated with outcome and AED response. The rationale of this study is that understanding the genetic contribution for outcome and AED response will translate into personalized medication choices, early identification of patients at risk for a more severe outcome, and elucidation of novel biological pathways for therapy development. The central hypothesis will be tested by pursuing two specific aims. As a first aim, this study will determine genetic factors associated with a similar longitudinal disease course. Preliminary data demonstrates that applying computational methods to determine the similarity of phenotypes enable the identification of novel genetic etiologies. This study will analyze EMR-derived longitudinal phenotypes in 2,500 individuals with available genetic data and identify genetic etiologies with related disease trajectories. As a second aim, this study aims to identify genetic factors that influence AED trajectories. AED response is not easily extracted from EMR datasets. However, longitudinal AED histories between patients can be compared, which may indicate biologically determined shared response patterns. This study will test whether patients with rare variants in shared genes have longitudinal AED trajectories that are more similar than expected by chance, highlighting empirical treatment patterns that may indicate gene-specific AED responses. This approach is innovative, as it leverages EMR as a ubiquitous, easily accessible, but previously unexamined data source in order to identify novel genetic risk factors in childhood epilepsies. The proposed research is significant as it is expected to expand understanding of genetic risk factors for outcome and AED response, which was previously not possible due to limited phenotypic data. This proposal extends the prior experience of the applicant to large, integrated cohort studies and provides invaluable training for a planned R01 application in in Year 3 of this award.